Update upstream source from tag 'upstream/2020.05.08'

[youtubedl] / youtube_dl / extractor / dctp.py

diff --git a/youtube_dl/extractor/dctp.py b/youtube_dl/extractor/dctp.py
index 3a6d0560e478cb08445f07d84578d28b3e4bc514..e700f8d86531415da0f1db0f2ccdaef6ea10ac53 100644 (file)
--- a/youtube_dl/extractor/dctp.py
+++ b/youtube_dl/extractor/dctp.py
@@ -5,82 +5,101 @@ from .common import InfoExtractor
 from ..compat import compat_str
 from ..utils import (
     float_or_none,
 from ..compat import compat_str
 from ..utils import (
     float_or_none,

-    unified_strdate,
+    int_or_none,
+    unified_timestamp,
+    url_or_none,

 )
 
 
 class DctpTvIE(InfoExtractor):
     _VALID_URL = r'https?://(?:www\.)?dctp\.tv/(?:#/)?filme/(?P<id>[^/?#&]+)'
 )
 
 
 class DctpTvIE(InfoExtractor):
     _VALID_URL = r'https?://(?:www\.)?dctp\.tv/(?:#/)?filme/(?P<id>[^/?#&]+)'

-    _TEST = {
+    _TESTS = [{
+        # 4x3

         'url': 'http://www.dctp.tv/filme/videoinstallation-fuer-eine-kaufhausfassade/',
         'url': 'http://www.dctp.tv/filme/videoinstallation-fuer-eine-kaufhausfassade/',

+        'md5': '3ffbd1556c3fe210724d7088fad723e3',

         'info_dict': {
             'id': '95eaa4f33dad413aa17b4ee613cccc6c',
             'display_id': 'videoinstallation-fuer-eine-kaufhausfassade',
         'info_dict': {
             'id': '95eaa4f33dad413aa17b4ee613cccc6c',
             'display_id': 'videoinstallation-fuer-eine-kaufhausfassade',

-            'ext': 'flv',
+            'ext': 'm4v',

             'title': 'Videoinstallation für eine Kaufhausfassade',
             'description': 'Kurzfilm',
             'title': 'Videoinstallation für eine Kaufhausfassade',
             'description': 'Kurzfilm',

-            'upload_date': '20110407',

             'thumbnail': r're:^https?://.*\.jpg$',
             'duration': 71.24,
             'thumbnail': r're:^https?://.*\.jpg$',
             'duration': 71.24,

+            'timestamp': 1302172322,
+            'upload_date': '20110407',

         },
         },

-        'params': {
-            # rtmp download
-            'skip_download': True,
-        },
-    }
+    }, {
+        # 16x9
+        'url': 'http://www.dctp.tv/filme/sind-youtuber-die-besseren-lehrer/',
+        'only_matching': True,
+    }]
+
+    _BASE_URL = 'http://dctp-ivms2-restapi.s3.amazonaws.com'

 
     def _real_extract(self, url):
         display_id = self._match_id(url)
 
 
     def _real_extract(self, url):
         display_id = self._match_id(url)
 

-        webpage = self._download_webpage(url, display_id)
+        version = self._download_json(
+            '%s/version.json' % self._BASE_URL, display_id,
+            'Downloading version JSON')

 
 

-        video_id = self._html_search_meta(
-            'DC.identifier', webpage, 'video id',
-            default=None) or self._search_regex(
-            r'id=["\']uuid[^>]+>([^<]+)<', webpage, 'video id')
+        restapi_base = '%s/%s/restapi' % (
+            self._BASE_URL, version['version_name'])

 
 

-        title = self._og_search_title(webpage)
+        info = self._download_json(
+            '%s/slugs/%s.json' % (restapi_base, display_id), display_id,
+            'Downloading video info JSON')

 
 

-        servers = self._download_json(
-            'http://www.dctp.tv/streaming_servers/', display_id,
-            note='Downloading server list', fatal=False)
+        media = self._download_json(
+            '%s/media/%s.json' % (restapi_base, compat_str(info['object_id'])),
+            display_id, 'Downloading media JSON')

 
 

-        if servers:
-            endpoint = next(
-                server['endpoint']
-                for server in servers
-                if isinstance(server.get('endpoint'), compat_str) and
-                'cloudfront' in server['endpoint'])
-        else:
-            endpoint = 'rtmpe://s2pqqn4u96e4j8.cloudfront.net/cfx/st/'
+        uuid = media['uuid']
+        title = media['title']
+        is_wide = media.get('is_wide')
+        formats = []

 
 

-        app = self._search_regex(
-            r'^rtmpe?://[^/]+/(?P<app>.*)$', endpoint, 'app')
+        def add_formats(suffix):
+            templ = 'https://%%s/%s_dctp_%s.m4v' % (uuid, suffix)
+            formats.extend([{
+                'format_id': 'hls-' + suffix,
+                'url': templ % 'cdn-segments.dctp.tv' + '/playlist.m3u8',
+                'protocol': 'm3u8_native',
+            }, {
+                'format_id': 's3-' + suffix,
+                'url': templ % 'completed-media.s3.amazonaws.com',
+            }, {
+                'format_id': 'http-' + suffix,
+                'url': templ % 'cdn-media.dctp.tv',
+            }])

 
 

-        formats = [{
-            'url': endpoint,
-            'app': app,
-            'play_path': 'mp4:%s_dctp_0500_4x3.m4v' % video_id,
-            'page_url': url,
-            'player_url': 'http://svm-prod-dctptv-static.s3.amazonaws.com/dctptv-relaunch2012-109.swf',
-            'ext': 'flv',
-        }]
+        add_formats('0500_' + ('16x9' if is_wide else '4x3'))
+        if is_wide:
+            add_formats('720p')

 
 

-        description = self._html_search_meta('DC.description', webpage)
-        upload_date = unified_strdate(
-            self._html_search_meta('DC.date.created', webpage))
-        thumbnail = self._og_search_thumbnail(webpage)
-        duration = float_or_none(self._search_regex(
-            r'id=["\']duration_in_ms[^+]>(\d+)', webpage, 'duration',
-            default=None), scale=1000)
+        thumbnails = []
+        images = media.get('images')
+        if isinstance(images, list):
+            for image in images:
+                if not isinstance(image, dict):
+                    continue
+                image_url = url_or_none(image.get('url'))
+                if not image_url:
+                    continue
+                thumbnails.append({
+                    'url': image_url,
+                    'width': int_or_none(image.get('width')),
+                    'height': int_or_none(image.get('height')),
+                })

 
         return {
 
         return {

-            'id': video_id,
+            'id': uuid,
+            'display_id': display_id,

             'title': title,
             'title': title,

+            'alt_title': media.get('subtitle'),
+            'description': media.get('description') or media.get('teaser'),
+            'timestamp': unified_timestamp(media.get('created')),
+            'duration': float_or_none(media.get('duration_in_ms'), scale=1000),
+            'thumbnails': thumbnails,

             'formats': formats,
             'formats': formats,

-            'display_id': display_id,
-            'description': description,
-            'upload_date': upload_date,
-            'thumbnail': thumbnail,
-            'duration': duration,

         }
         }